Other Parts Discussed in Thread: DP83848C, DP83869
I'm connecting an ethernet switch to TI's PHY (DP83869HMRGZT) on the same board very close together, and I had planned to use capacitive coupling for 1000 and 100 Mbps operation.
Per TI's E2E FAQ capacitive coupling should be supported: https://e2e.ti.com/support/interface-group/interface/f/interface-forum/1265741/faq-ti-ethernet-phy-capacitive-coupling-transformerless-operation?tisearch=e2e-sitesearch&keymatch=capacitive%20coupling
My question: Regarding that FAQ, are there any limitations in using DP83869HM with capacitive coupling? Specifically, will DP83869HM function correctly when capacitively-coupled with auto-negotiation 1000/100 advertised (no 10BASE-T necessary)?
I haven't found any TI documentation on capacitive coupling with 1000Mbps specifically. The FAQ only mentions 10/100 and links an app note for DP83848, which is a different 10/100-only PHY https://www.ti.com/lit/SNLA088A. It states: "Therefore, it is recommended that transformerless operation be restricted to forced 100Mb/s modes, with autonegotiation disabled".
The app note reasoning for forced 100Mbps seems to be that 10BASE-T wouldn't be symmetrical with capacitive coupling.
Another E2E response verifies that 10/100 auto-negotiation may have issues for DP83848C due to pulse distortion: https://e2e.ti.com/support/interface-group/interface/f/interface-forum/1104097/dp83848c-transformerless-capacitively-coupled-ethernet
Microchip's guidance on the ethernet switch capacitive coupling states that 1000Mbps needs to be treated differently, "Auto-negotiation must be enabled for 1000BASE-T connections. This is different from 10BASE-T and 100BASE-TX transformerless connections, where it is sometimes suggested to disable auto-negotiation for transformerless applications."
https://microchip.my.site.com/s/article/Capacitive-Coupling-for-Gigabit-Ethernet
All that leads me to believe that 1000BASE-T and 100BASE-TX should be okay capacitively coupled with auto-negotiation, but 10BASE-T may have issues with capacitive coupling due to asymmetrical operation. Please let me know if I'm mistaken.
Thanks!
